Creating new drugs is an important national need, but a critical barrier is escalating cost, often due to late failures in the pipeline. Although in vitr screens targeting proteins or cell lines provide important avenues, in vivo screens on intact vertebrates can help overcome late-failure problems. Intact animal disease models provide an unbiased approach that screens all proteins and tissues, many of which may not yet be known to play a role in disease. The goal of this resource-related research project is to develop animal models and related materials for using global gene expression patterns to identify novel therapies for human disease. Larval fish provide especially useful subjects for intact-animal drug screens because they have organs that function like human organs, including epithelia for absorption, livers for metabolism, kidneys for excretion, and organs like eyes, ears, brain and thyroid sensitive to drug toxicity. Fish can absorb drugs directly from the water and their small size makes testing economical. Disease states are characterized by constellations of gene activity indicating disease etiology and response to disease. In an innovative screen, this project will identify transcriptional disease signatures (TDSs), suites of genes whose activities change in disease, and will develop efficient methods to identify compounds that return the TDS to that of healthy fish. The project predicts that TDS changes will reveal drug effects earlier than altered morphologies would and will represent a whole-organism response to therapy better than in vitro protein- or cell-based tests. Outcomes include improved screens for chronic disease therapeutics and improved animal models for drug screening. Investigations will use two disparate models for human disease: a transgenic medaka model for malignant melanoma (MM, a deadly human skin cancer), and a mutant zebrafish model for Fanconi anemia (FA, a disease of bone marrow failure that inhibits stem cell proliferation and survival). Aim 1 is to use RNA-seq to identify TDSs that define diseased vs. healthy fish. Aim 2 is to evaluate HTG-Edge and NanoString nCounter for their efficiency in assaying TDSs. Aims 3 and 4 are to conduct pilot screens on MM medaka (Aim 3) and FA zebrafish (Aim 4) to identify compounds that change the TDS from disease to healthy profiles. 'Hits' will be candidates for mechanistic analyses in fish and for safety and efficacy studies in mammals suffering MM and FA disease. Achieving these aims will have an enduring influence on translational research by developing novel but widely applicable drug screen resources and methodologies, as well as improving animal models of human disease and exploiting the advantages of intact-animal screens, thus complementing other screen systems. Results will enhance research infrastructure by outlining best practice protocols for using the innovative concept of transcriptome disease signatures to identify potential therapeutics for human disease and enhance translational research by providing lead compounds for chemical therapies for melanoma and Fanconi anemia.